检索结果-内蒙古大学图书馆

3rd International Conference on Electrical, computer and Communication engineering, ECCE 2023

作者： Abrar, Raiyan Islam, Md. Tariquel Naim, Mahapara Tabassum, Nuzhat Military Institute of Science and Technology Dept. of Computer Science and Engineering Dhaka Bangladesh

ISBN: (纸本)9798350345360

3D modeling in the shoe industry is growing as the technology becomes more widely available. 3D modeling lets companies customize shoes for individual customers, making manufacturing more flexible and efficient. In the shoe industry of Bangladesh, using smartphone images for 3D modeling is a fairly new idea. This allows designers and manufacturers to collaborate and quickly share designs with customers for feedback. A photogrammetry pipeline is a cost-effective and efficient way to create 3D models. This study proposes a potential and affordable method for using photogrammetry to create 3D representations of footwear. The major goal was to determine whether photogrammetry, as opposed to RGB depth camera photos, can more conveniently construct 3D models of footwear from a small number of smartphone images. In order to assure correct reconstruction, the pipeline required taking a collection of photos from 6-9 distinct perspectives and augmenting the dataset. Then, from these images, point clouds are produced, and from these aligned point clouds, the 3D mesh is produced. The mesh is then textured and smoothed to give color. In Bangladesh's shoe industry, this 3D modeling pipeline can be used for AR shoe try-on. © 2023 IEEE.

关键词： 3D modeling

来源：评论

学校读者我要写书评

暂无评论

Federated Multiarmed Bandits Under Byzantine Attacks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第6期6卷 1488-1501页

作者： Saday, Artun Demirel, Ilker Yildirim, Yigit Tekin, Cem Bilkent University Department of Electrical and Electronics Engineering Ankara06800 Turkey Massachusetts Institute of Technology Computer Science & Artificial Intelligence Laboratory CambridgeMA02139 United States

Multiarmed bandits (MAB) is a sequential decision-making model in which the learner controls the trade-off between exploration and exploitation to maximize its cumulative reward. Federated multiarmed bandits (FMAB) is an emerging framework where a cohort of learners with heterogeneous local models play an MAB game and communicate their aggregated feedback to a server to learn a globally optimal arm. Two key hurdles in FMAB are communication-efficient learning and resilience to adversarial attacks. To address these issues, we study the FMAB problem in the presence of Byzantine clients who can send false model updates threatening the learning process. We analyze the sample complexity and the regret of β-optimal arm identification. We borrow tools from robust statistics and propose a median-of-means (MoM)-based online algorithm, Fed-MoM-UCB, to cope with Byzantine clients. In particular, we show that if the Byzantine clients constitute less than half of the cohort, the cumulative regret with respect to \beta-optimal arms is bounded over time with high probability, showcasing both communication efficiency and Byzantine resilience. We analyze the interplay between the algorithm parameters, a discernibility margin, regret, communication cost, and the arms’ suboptimality gaps. We demonstrate Fed-MoM-UCB's effectiveness against the baselines in the presence of Byzantine attacks via experiments. © 2020 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Transfer Learning Models for Detecting Autism Spectrum Disorder Using Facial Images 11

Performance Analysis of Transfer Learning Models for Detecti...

引用

11th International Conference on Intelligent Systems and Embedded Design, ISED 2023

作者： Sharma, Harendra Sharma, Oshin Srm Institute of Science and Technology Department of Computer Science and Engineering Modinagar Ghaziabad India

ISBN: (纸本)9798350318159

Autism Spectrum Disorder (ASD) is a type of disorder which impacts how individuals communicate interact socially and behave. It is crucial to detect and diagnose ASD on for intervention and support. Over the years advanced deep learning models have shown great potential in various medical fields, including detecting autism. The face of a person can serve as a biomarker because it can reflect the state of the brain making it a convenient and straightforward tool, for diagnosis. In this study we employ learning methods based on convolutional neural networks (CNNs) to identify children with autism by analyzing their facial images. This research paper examines three learning models. Xception, VGG16 and VGG19. After conducting training and validation using the optimized configuration the Xception model showcases the level of performance, with an accuracy rate of 89% on test dataset. On the hand models, like VGG16 and VGG19 achieved accuracies of 76% and 80% respectively. To compare their effectiveness, in detecting autism using data. The study evaluates these models performance based on accuracy, sensitivity, specificity and computational efficiency. © 2023 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Enhanced perceptual wavelet packet features for spontaneous Kannada sentence recognition under uncontrolled conditions

引用

International Journal of Speech technology 2025年第1期28卷 153-174页

作者： Shanthamallappa, Mahadevaswamy Pradeep Kumar, B.P. Department of Electronics and Communication Engineering Vidyavardhaka College of Engineering Mysuru Visvesvaraya Technological University Karnataka India Department of Computer Science and Design Atria Institute Of Technology Bengaluru Visvesvaraya Technological University Karnataka India

A Robust Automatic Speech Recognition (ASR) system is proposed through a hybrid combination of Perceptual Wavelet Packet features, Deep Neural Network-Hidden Markov Model (DNN-HMM) acoustic models, and n-gram language models for recognizing Spontaneously spoken Kannada sentences. Two Experiments are performed on own Kannada speech corpus and the widely recognized TIMIT corpus. First one is the experiment on the clean speech datasets and the second one is on the speech degraded by noise data chosen from Noise-US dataset. The proposed system performance is evaluated through the metric Word Error Rate (WER). A comparative study on the performance analysis is presented. The results of the experimental investigations achieved an average reduction in Word Error Rate (WER) of 1.8% over Mel Frequency Cepstral Coefficients (MFCCs) and 2.1% over Perceptual Linear Prediction (PLP) features. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

Blockchain-Based Paper Review System 9

Blockchain-Based Paper Review System

引用

9th International Conference on Smart Computing and Communications, ICSCC 2023

作者： Peter, Allen Xavier Hassan, Sahil Sudev Krishnan, K. Jose, Tony Rakhee, M. Muthoot Institute of Technology and Science Dept. of Computer Science and Engineering Kochi India

ISBN: (纸本)9798350314090

Paper review and publishing is an important part of a researcher's life. It is where a researcher's time and resources that he has spent bears fruit. And this is enabled by getting his work reviewed by different peers. In our current state of things, researchers face a lot of issues while they try to publish their research manuscript. These issues include quality of peer-reviews, rise of predatory journals, bias between reviewers, and other issues regarding academic integrity. These issues in the traditional system are what make the system inefficient. Therefore, we propose a more efficient blockchain-based system that will try to tackle these issues. Hyperledger Fabric is one of the most popular Blockchain platforms which provides a permissioned blockchain. Our proposed system would leverage the permissioned blockchain. This is to ensure that not everyone from outside the system has access to the data present in the system and only the members of the blockchain are able to access the data. The workflow of the peer-review system would be stored as transactions. The blockchain would be used to keep track of these transactions and order them properly in order to maintain provenance. This system would prove to be a lot more efficient than the traditional existing systems. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection System in Internet of Things-Empowered Cybersecurity by Using Various Algorithms 2

Intrusion Detection System in Internet of Things-Empowered C...

引用

2nd IEEE International Conference on Data science and Network Security, ICDSNS 2024

作者： Surisetti, Sivaleela Bhavanam, S. Nagakishore Oruganti, Vijageesh Devi, M Vasuja Acharya Nagarjuna University Department of Computer Science & Engineering Andhra Pradesh Guntur India Department of Computer Science & Engineering Jabalpur Barela India Vit University Department of Computer Science & Engineering Vellore Institute of Technology Andhra Pradesh India

ISBN: (纸本)9798350373110

Internet of Things (loT) is the new expertise for various significant application growth. To avoid adversarial attacks, network intrusion, and fraud, Intrusion Detection System (IDS) is one of a primary element of organization. An IDS in loT -empowered cybersecurity observes network traffic to alert and detect malicious activities or policy violations. It secures the devices of loT and networks by recognizing unauthorized access or anomalies. Recently, various researchers have identified significant ways for IDS by utilizing Artificial Intelligence (AI) techniques of Machine Learning (ML) and Deep Learning (DL) algorithms. This research represents several methodologies like Artificial Neural Network (ANN), K- Nearest Neighbor (KNN), Decision Tree (DT), Bagging, Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), Deep Neural Network (DNN), and Hybrid Deep Belief Network (HDBN) that are employed for IDS. Accuracy, precision, sensitivity, specificity, and so on, are utilized as the parameters in this study. © 2024 IEEE.

关键词： Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

Handwritten Prescription Recognition Using VGG Based Architecture with Bi-LSTM

Handwritten Prescription Recognition Using VGG Based Archite...

引用

2024 International Research Conference on Smart Computing and Systems engineering, SCSE 2024

作者： Kavinda, Isuru Fernando, Harinda Sri Lanka Institute of Information Technology Department of Computer Science and Software Engineering Malabe Sri Lanka Sri Lanka Institute of Information Technology Department of Computer System Engineering Malabe Sri Lanka

ISBN: (纸本)9798350375688

Illegible handwriting on medical prescriptions poses a significant challenge, often leading to the misinterpretation of drug names and dosages. This issue primarily stems from doctors' use of Latin abbreviations, driven by time constraints during patient consultations. The resultant discrepancies in prescription interpretation can potentially expose patients to incorrect medications and dosages, leading to adverse effects. In response to this concern, this study proposes a novel system integrating Convolution Neural Network (CNN) and Recurrent Neural Network (RNN) models. Promising findings are reported in this study, achieving a loss of 0.4874 and an accuracy of 0.83 in the latest trial. The final model offers a potential solution to the persistent problem of illegible prescriptions, enhancing patient safety in the healthcare domain. © 2024 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Exploring Machine learning applications in cyber attacks detection

Exploring Machine learning applications in cyber attacks det...

引用

2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024

作者： Rai, Rina Mishra, Aditi Singh, Divya Kumar, Gaurav Kumar, Anuj Arya, Bhupal Madan Mohan Malaviya University of Technology Department of Computer Science Gorakhpur India Parul University Vadodara Department of Computer Science Gujarat Vadodara India University of Lucknow Roorkee Institute of Technology Department of Computer Science Lucknow India Uttaranchal Institute of Technology Department of Computer Science and Engineering Dehradun India Babasaheb Bhimrao Ambedkar University Department of Information Technology Lucknow India RIT Department of Computer Science and Engineering Roorkee India

ISBN: (纸本)9798350378092

Attackers and hackers employ many tactic known as phishing, cyber-attacks, dos attacks DOS attacks, man in middle attacks etc to damage trustworthy URLs as well as important websites. In this research, we discussed various cyber threats, the interrelation of cyber threats, and machine learning types of phishing attacks various techniques to detect cyber phishing by using machine learning, and we have also compared various phishing attacks with a few machine learning techniques with the help of a classification technique like supervised learning in which we apply SVM, random forest and extreme learning machine, the NSL-knowledge discovery is used in the evaluation of intrusion detection method also we have compared the accuracy in which precision-recall is used to find the accuracy and to calculate the performance and we concluded that ELM method give best result in terms of accuracy. © 2024 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

Integration of AI in the Realm of Software Development

Integration of AI in the Realm of Software Development

引用

2023 International Conference on Advances in Computation, Communication and Information technology, ICAICCIT 2023

作者： Bhuvana, J. Ranjan, Vivek Bhaduariya, Nirmednra Department of Computer Science and Information Technology Bangalore India Noida Institute of Engineering and Technology Department of Computer Science Uttar Pradesh Greater Noida India Tula's Institute Department of Computer Science and Engineering Dehradun India

ISBN: (纸本)9798350344387

Software engineering has increasingly utilized "Artificial Intelligence"(AI) to improve self-organizing IT solutions. This research aims to fill the existing void in the thorough examination of artificial intelligence's (AI) impact on software engineering. To achieve this, the study used a combination of an extensive literature review and qualitative Conversations with software engineers. The objective of the study is to evaluate the present status of AI integration, possible progress, and the accompanying hazards in the realm of software development. The results demonstrate that artificial intelligence has made substantial progress in multiple domains. By employing algorithmic methodologies, it has automated time-consuming chores such as debugging and documentation, resulting in a more efficient software development and testing process. AI has facilitated the methodical examination of extensive datasets, detecting patterns and novel clusters of information. Moreover, the utilization of neural networks has enabled comprehensive analysis of these datasets. Hence, AI expedites progress, diminishes development expenses, and augments total efficacy in software engineering. © 2023 IEEE.

关键词： AI DL ML SDLC and pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Customer Spending Prediction on E-Commerce Website 15

Customer Spending Prediction on E-Commerce Website

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Salake, Sudha V. Patil, Pankaja Kunchur, Pavan Bangarashetti, Sadhana Managuli, Manjunath Patil, Sheetal B. Department of Computer Science and Engg KLS Gogte Institute of Technology Belagavi India Department of Information Science and Engg Basaweshwar Engineering College Bagalkot India Department of Electronics and Communication Engineering KLS Gogte Institute of Technology Belagavi India Department of Computer Science and Engg SJB Institute of Technology Belagavi India

ISBN: (纸本)9798331300579

In the ever-evolving world of e-commerce, understanding customer behavior and predicting their spending patterns is crucial for businesses to thrive. This work presents an innovative application designed to predict customer spending on an e-commerce website. By leveraging advanced machine learning techniques and utilizing comprehensive customer data, the application offers valuable insights that can be used to optimize sales strategies and enhance customer *** application employs a robust predictive model trained on historical customer data, including purchase history, browsing patterns, demographic information, and transactional *** application incorporates real-time data updates to ensure continuous improvement and adaptability to changing customer *** proposed application offers a powerful tool for e-commerce businesses to predict customer spending, optimize sales strategies, and deliver personalized experiences. By harnessing the power of machine learning and leveraging comprehensive customer data, businesses can stay competitive in the dynamic e-commerce landscape and foster long-term customer loyalty. © Grenze Scientific Society, 2024.

关键词： Marketplaces

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：